Jump to content
IGNORED

Recent Airport Express toslink jitter


Recommended Posts

Hello,

 

I've heard here that recent AE is somehow wrong for music streaming because of its poorer jitter in optical output. Where can I find any details about that, is it proven?

 

I do remember there was a stereofile's tests where first AEs were bit to bit perfect. If it's false now with newest AEs, that makes trouble... how to find now which one is "old" model, which is "new", what is criteria? And what really happens now with jitter?

 

Second: does anyone heard of any AE power source tweaks, from original impulse to external "clear linear", does it make sense in part of optical jitter reduce? Is it necessary?

 

Any info appreciated.

 

Vit

 

Link to comment

Well, you are right of course. This is one of those things that becomes 'fact' by virtue of the number of people that repeat it. There are many such examples, on this and other audio forums, where facts become established by this manner!

 

The plain fact of the matter, of course, is that the vast majority of said experts wouldn't know what jitter sounded like, or what detrimental effects it was having, if it bit them in the crotchets! Jitter is the darling of audiophile 'experts' everywhere for this very reason.

 

If you fancy an AE and it suits your setup to install one, then go ahead and have a listen to one. If it sounds good to you then you're done. The same goes for power tweaks and the like. If you can identify a problem, as you sit and listen to your music, then you have something to fix. If it sounds as though it could be better, then you have some research to do. If it sounds brilliant then you have some happy nights ahead of you.

 

My advice, for what it's worth, is if you need to 'fix' a bit of kit - by pimping it's power supply for example - then go buy a better bit of kit!

 

Have fun.

 

Link to comment

Thank you Bob.

 

I'm absolutely satisfied by AE. I was confused that most complains about AE are from people who are selling so called jitter minimizers, they write everywhere about it's terrible jitter, then explain what jitter is, and so get you interested in their stuff. This is marketing, okay. I was wondering is there something behind it instead of get you involved and impressed by "another magic boxes", or not. Facts, I mean.

 

Another reason, I'm trying to prove for some friends who are kind of old-schoolars, that computer can sound equal to CDs that could push you from your room space by its amount in shelves...

 

Link to comment

I am not selling anything, and if you are satisified with the sound you are getting from the AE, great.

BUT: the digital stream from the AE, (and from the digital output of a Mac computer as well) is very high in jitter, and this will effect the sonic performance. Jitter, has nothing to do with being bit perfect (they are two different attributes of digital music streaming) and as far as I know, the AE when properly configured is bit perfect.

But the jitter is high, in the area of 1000s of pS. For reference, the best asynchronous computer/DAC interfaces achieve jitter levels in the double digits of pS: around 20-60 pS, so as you can see, the difference is quite large.

If you are concerned with getting the best sound possible, the AE is not the way to go, but if you are happy as is, then that is of course, fine.

 

SO/ROON/HQPe: DSD 512-Sonore opticalModuleDeluxe-Signature Rendu optical with Well Tempered Clock--DIY DSC-2 DAC with SC Pure Clock--DIY Purifi Amplifier-Focus Audio FS888 speakers-JL E 112 sub-Nordost Tyr USB, DIY EventHorizon AC cables, Iconoclast XLR & speaker cables, Synergistic Purple Fuses, Spacetime system clarifiers.  ISOAcoustics Oreas footers.                                                       

                                                                                           SONORE computer audio

Link to comment

this is what I mean by the invention of facts. :)

 

What jitter is it that is being measured? What frequencies are being measured? What level of interface jitter is known to be 'lethal' to the receiving DAC? Are all DACs incapabale of coping with this jitter? What is the audible effect of 1000ps of interface jitter?

 

Measurements are fine, but they need to be explained within some kind of context. How would you indentify that a system was suffering from an excessively high interface jitter. What would you listen out for? Would you always be able to hear it in a system using an optical interface?

 

The AE may well not be the pinnacle of audio excellence but that could have as much to do with any number of 'in chain' events as it has to do with jitter. On the other hand, if, as youmsay, an AE is able to transmit bit perfect audio data, then it may well be as good as any other item performing the same function.

 

Your mileage probably does vary. :)

 

Link to comment

that have been done:

 

"What jitter is it that is being measured? What frequencies are being measured? What level of interface jitter is known to be 'lethal' to the receiving DAC? Are all DACs incapabale of coping with this jitter? What is the audible effect of 1000ps of interface jitter?"

 

Correspond precisely with my listening impressions, both of the toslink output of the AE and of my MacBook. I achieve a huge increase in quality by using a very low jitter, asynchronous, USB-SPDIF converter. This is not rocket science here, and neither are these kind of results unusual, they are commonplace to the many people who have experienced them.

Additionally, the term "jitter rejection" in terms of the DAC being used is utter nonsense in my experience. Every DAC sounds better when fed a very low jitter data stream. For example, my DAC uses the ESS 9018 chip with its onboard ASRC engaged. This topology is acknowledged to achieve excellent "jitter rejection", but still, the DAC only sounds its best when fed from very low jitter source.

The toslink output from the AE and/or Mac achieve reasonable, but not what I would consider "high end" performance. For example, a simple DVD player (Oppo), feeding the same DAC via SPDIF with a decent cable will be superior sounding to the toslink feed from an AE or Mac.

 

 

 

SO/ROON/HQPe: DSD 512-Sonore opticalModuleDeluxe-Signature Rendu optical with Well Tempered Clock--DIY DSC-2 DAC with SC Pure Clock--DIY Purifi Amplifier-Focus Audio FS888 speakers-JL E 112 sub-Nordost Tyr USB, DIY EventHorizon AC cables, Iconoclast XLR & speaker cables, Synergistic Purple Fuses, Spacetime system clarifiers.  ISOAcoustics Oreas footers.                                                       

                                                                                           SONORE computer audio

Link to comment

For example, my DAC uses the ESS 9018 chip with its onboard ASRC engaged. This topology is acknowledged to achieve excellent "jitter rejection", but still, the DAC only sounds its best when fed from very low jitter source.

 

It's not that much DAC chip, but it's the S/PDIF receiver who needs to deal with this. That's where the PLL is. Even when the suppression is doubled with ASRC, there's still the ASRC rate control speed != 0.

 

Naturally minimizing amount of jitter to deal with in first place is best situation. That's why DACs running out of their own steady clocks are best, be it async USB/FW interface or PCI(e) card.

 

However, when there still are cases where jitter exists on the S/PDIF signal, there are differences between DACs (as in devices) on their ability to minimize the impact of jitter. Functionally AE is excellent device and at the moment not easily replaceable with something else, thus I think it's still valid to concern cases where best performance can be achieved with it.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

 

Yes, sort of... It appears that you are not familiar with the ESS 9018 topology. All the components are on the single DAC chip: SPDIF receiver, DPLL, and ASRC-so the "jitter rejection" all occurs on the single chip.

 

I agree that some DACs are still more sensitive to incoming jitter than others, but I have never used a DAC that did not sound better when supplied a low jitter signal.

In terms of "jitter rejection" (or "jitter immunity") I am putting these phrases in quotes because they are often used in the marketing materials of DACs-mostly I am referring to those materials, as they give the impression (wrongly) to the consumer that source jitter is irrelevant to sonic performance.

 

 

 

 

 

SO/ROON/HQPe: DSD 512-Sonore opticalModuleDeluxe-Signature Rendu optical with Well Tempered Clock--DIY DSC-2 DAC with SC Pure Clock--DIY Purifi Amplifier-Focus Audio FS888 speakers-JL E 112 sub-Nordost Tyr USB, DIY EventHorizon AC cables, Iconoclast XLR & speaker cables, Synergistic Purple Fuses, Spacetime system clarifiers.  ISOAcoustics Oreas footers.                                                       

                                                                                           SONORE computer audio

Link to comment

It appears that you are not familiar with the ESS 9018 topology. All the components are on the single DAC chip: SPDIF receiver, DPLL, and ASRC-so the "jitter rejection" all occurs on the single chip.

 

True, they've been a little bit too secretive, no datasheets or anything available, for me to be really interested. That single chip approach probably makes it quite sensitive for domain leakage, especially for capacitive one.

 

as they give the impression (wrongly) to the consumer that source jitter is irrelevant to sonic performance

 

Source-clocking is always suspect to frequency variations.

 

And if we go scientific, there's no such thing as jitter-free clock, it's all just matter of measurement precision.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

I have an airport express and absolutely love it. I love the incredible convenience it gives me in the ability to play all my iTunes music from my computer that is far, far away. It sounds very good too. My airport G model (not the newer N model) is optically feeding a Matrix Mini I and I do enjoy bit perfect music coming out of it.

 

Bob, I think you may believe that once a device is bit perfect than that's it. If you've been here on this forum a while you will soon find that bit perfect is only the beginning.

 

I have been able to verify that bit perfect data can sound different through different devices in my own system. I've taken AIFF files and played them thru iTunes on My G4 music server serving the airport, or CoolEdit on my Windows XP digital audio workstation with an Echo Mia soundcard, and finally through my standard CD/DVD player by burning a CD. All these devices are connected to my Matrix Mini, the airport via toslink and the XP and CD player thru co-ax.

 

Unsurprisingly, the airport is my least favorable connection. It doesn't sound bad, just not as good as the others. Imaging is a bit more vague and less coherent relative to the other connections. Probably all due to its high jitter. Perhaps the type of connections pays a part as well, but I don't the difference between co-ax and toslink would be as significant compared to what I heard. They are both SPDIF after all.

 

My plan is to get a jitter busting device to finally round out my system. There are so few out there, though.

 

CD

 

 

Link to comment

My G4 733 (running 10.4.11 Tiger) choked when I upgraded to iTunes 9. Choke may be a strong word. It played the files fine through my airport express, but the moment I tried to do something like invoke the graphic EQ or get info on a musical file, the server choked. If I left it alone it was OK and sounded good.

 

Not being totally comfortable with that situation and recently having been handed down a Dual G4 1.3 Ghz machine, I rebuilt and transferred my music server to that machine. This server (running 10.5.8 Leopard) plays fine and doesn't blink.

 

Interestingly, when my PowerMac G4 733 was choking, I launched the activity monitor to try to see where the bottleneck was. It gave nothing away. CPU was fine at around 10 to 12 %, disk activity was nothing special, network usage, nothing. The server was choking and activity monitor gave me no clue where the problem was.

 

CD

 

EDIT: added OS details

 

Link to comment

Computer or AE - you need a DAC with excellent jitter suppression if you want top quality output.

 

Benchmark DAC 1 is great value and has a string reputation for being immune to jitter.

 

Jitter is audible, even at reasonably low levels of 258 pico-secs - provided it is non-random. Slightly less clarity in the mids and treble is the usual manifestation. You probably won't hear it on anything but a high quality system and even then the differences are small (but worth addressing if you are an audiophile)

 

Link to comment

on the contrary, it is because I have been around here long enough, that I happen to believe that none of it is 'important' in itself. Bit perfect, jitter numbers, spdif vs usb vs firewire, it's all meaningless armchair twaddle unless it's tied into the real world.

 

It's the lack of anything meaningful that sets my teeth on edge. We are all, I assume, trying to assemble systems we find enjoyable. Perfection, whether it be software or hardware related, has got little to do with that goal unless that very perfection 'is' your goal. Oh, hang on, I'm off on one here.

 

Right, my point is that your opinion of the AE, as being a bit vague in the soundstage and somewhat less coherant, is a valid and useful piece of information the OP can use. It is real world information that is helpful. Whether or not it has 1000ps or 258ps, of some jitter or other, is useless information that is not in the least helpful.

 

Numbers say nothing, that's my point, really.

 

Link to comment

I agree that how something sounds should be the ultimate arbiter of our audio choices. But:

 

"Right, my point is that your opinion of the AE, as being a bit vague in the soundstage and somewhat less coherant, is a valid and useful piece of information the OP can use. It is real world information that is helpful. Whether or not it has 1000ps or 258ps, of some jitter or other, is useless information that is not in the least helpful."

 

Neither jitter measurements are useful. Especially in this case, where we know the jitter of both the AE, and the optical output of a Mac computer are an order of magnitude higher (not just a few pS, but 10-100x more) than that of a good asychronous interface. I have listened, and I know how jitter manifests itself in playback degradation. No one should consider only measurements when evaluating a component, but jitter numbers this high are bad, period. and the listening results just confirm that analysis.

I have nothing against the AE, it is a cool little wireless interface, which packs a huge amount of functionality in a convenient small and affordable package, my only point regarding it is that it does not allow for the same performance level as good asynchronous computer interfaces-it is a compromise solution.

 

 

 

SO/ROON/HQPe: DSD 512-Sonore opticalModuleDeluxe-Signature Rendu optical with Well Tempered Clock--DIY DSC-2 DAC with SC Pure Clock--DIY Purifi Amplifier-Focus Audio FS888 speakers-JL E 112 sub-Nordost Tyr USB, DIY EventHorizon AC cables, Iconoclast XLR & speaker cables, Synergistic Purple Fuses, Spacetime system clarifiers.  ISOAcoustics Oreas footers.                                                       

                                                                                           SONORE computer audio

Link to comment

I wonder what is overall system level where you can say: here AE is perfect, and here AE became a compromise.

 

Those who modded AE were complaining first that AE comparing to (looks like) a very good transports is step back in quality. Modding, they finally got AE equal to transports. Hard to say how good their ears and systems, but at least their success drives me. I would really like to skip transport story and move to useful wi-fi solution, and I know for sure that my system is not the very "end", however I can hear transport vs AE difference. Not dramatic, though. My hope was to skip wires, strange tales of $$$ jitter rejectors, and get AE better than it is, as I see potential for this and reasonable success stories.

 

Sure, there are systems where AE sounds terrible. Even modded, it will not achieve truly hi-end (sorry for this term) level, but where is the "break" level, where the system prove it's bad?

 

Link to comment

Here's a metric: Even cheap transports crush the AE. Not even close.

 

I modded the AE years ago - it was called the Off-Ramp WiFi. It had improved power supply, power mods, Superclock4 and output S/PDIF coax. Even with all of this, it was still pretty poor. Its due to the TI chip they use. No help possible. I sold one and then discontinued the product. A reclocker is a much better solution. AE is too broken.

 

Steve N.

Empirical Audio

 

Link to comment

Steve N:

 

Wrong. I've heard on few very unique systems that AE was close to good transports. Not equal, not miles away - close.

 

I'm sorry, don't you think you promote your magic reclockers too aggressive? Everywhere. Too much. You modded AE long time ago and that time it was great: you were selling it modded, more than one, for thousand bucks (wow!), and now you say that "it was still pretty poor"? Really? That's amazing.

Most funny: recent AE-n produced for years already doesn't have TI chip, is it surprise? Sure not, then why don't you point on this fact and discuss AE-n as an option, instead of focusing again and again and again on your wonderful boxes?

 

Link to comment

At the end of the day it's up to you to decide if you're happy with the sound.

 

I use one in a secondary headphone system. It feeds a $100 TC-7510 that then goes to a headphone amp. I control it all in my comfy chair using my iPod Touch. It's provided hours of musical enjoyment.

 

Bill

 

 

Simplicity is the ultimate sophistication.

Mac Mini->Roon + Tidal->KEF LS50W

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...